PROJECT SUMMARY/ABSTRACT Type 1 diabetes (T1D) is a devastating and chronic autoimmune disease that affects 0.5% of the US population. Strategies for targeted induction of immune tolerance promise to produce a durable cure for T1D. However, these approaches rely on knowledge of specific molecular targets of autoimmunity. INS, GAD2, and other known autoantigens are the current gold-standard biomarkers for T1D, but the prevalence of ?autoantibody-negative? sporadic and immunotherapy-associated cases suggests that there are additional, unknown causal antigens in T1D. Thus, high-throughput antigen discovery represents a critical next step to enable antigen-specific re- establishment of immune tolerance in T1D. In preliminary work, I adapted a proteome-wide phage display assay to identify antigenic targets of human serum autoantibodies. I applied this approach to sera from a cohort of patients with Autoimmune Polyendocrine Syndrome Type 1 (APS1), with the rationale that (i) APS1 patients have a high incidence of T1D and (ii) APS1 patients make high affinity antibodies, increasing the sensitivity of my assay to discover novel autoantigens. Strikingly, I found that many APS1 patients exhibited a strong autoantibody response to RFX6. As RFX6 is expressed specifically in pancreatic islets and has been implicated in congenital cases of diabetes and intestinal pathology, this discovery motivates a mechanistic investigation of the role of RFX6 in pancreatic autoimmunity. My central hypothesis is that RFX6 autoimmunity is sufficient to cause diabetes and that RFX6 is a previously unknown autoantigen in sporadic T1D. In my first aim, I will use mouse models to test the causal relationship between anti-RFX6 immune response and pancreatic pathology, including overt diabetes. In my second aim, I will determine whether T1D patients mount an immune response to RFX6. Specifically, I will test for RFX6 autoantibodies in sporadic T1D patients, as well as in autoantibody- negative and immunotherapy-induced diabetes where there is a pressing need for antigen discovery. Finally, I found in preliminary analyses that RFX6 may exhibit non-canonical regulation in the thymus: the pattern of RFX6 expression does not match that of classic tissue-specific autoantigens in the thymic epithelium. Therefore, in my third aim, I will dissect the pattern and regulation of thymic RFX6 expression. Results from this aim will have broad implications for the immunobiology of central tolerance, and more specifically, may elucidate mechanisms of loss of immune tolerance in the setting of autoimmune diabetes. Together, these aims will definitively determine the role and spectrum of immune response to RFX6. In the longer term, the goals of this project are to establish an actionable clinical biomarker, as well as to expand the autoantigen spectrum in T1D to enable improved specificity of future antigen-specific therapeutic approaches.